Articles |
From the Department of Anatomy and Embryology, University of Leiden, Netherlands (J.S., A.C.G. de G., J.C. van M.); Laboratoire de Physiopathologie du Développement, CNRS (URA 1337) et Ecole Normale Supérieure, Paris, France (M.A.G., V.E.K.); Department of Pathology, General Hospital Middelheim, Antwerp, Belgium (M.M.K.); and Department of Pathology, University of Washington, Seattle (S.M.S.).
Correspondence to Prof Dr A.C. Gittenberger-de Groot, Department of Anatomy and Embryology, University of Leiden, PO Box 9602, 2300 RC Leiden, The Netherlands. E-mail acgitten{at}rullf2.leidenuniv.nl
| Abstract |
|---|
|
|
|---|
-smooth muscle actin, smooth muscle myosin, calponin, and heavy caldesmon, and the expression of the membrane-related protein smooth muscle phosphoglucomutaserelated protein increase, whereas the expression of other proteins, such as fibronectin splice variants with extradomains A (EDA) and B (EDB), decreases. In this study, we investigated the differentiation of the SMCs of the ductus arteriosus during the development of intimal thickening. Ascending and descending aortas of the same age were used for comparison because these vessels lack intimal thickening. In the fetal ductus arteriosus, a relatively early differentiation of the contractile apparatus was observed compared with the ascending and descending aortas. EDA and EDB expression was already low, being similar in the ductus and descending aorta and even lower in the ascending aorta. In the neonatal ductus, SMCs of the media and outer intima were well differentiated and comparable with SMCs of the ascending aorta. Dedifferentiated SMCs, with a low expression of cytoskeletal and contractile proteins and a high expression of EDA and EDB, were found in regions in the inner intima that show features of progression of intimal thickening and in areas of cytolytic necrosis in the media. With a technique using in situ end labeling of DNA fragments, we found extensive apoptosis in the area of cytolytic necrosis and to a lesser extent in these areas of the inner intima. In conclusion, SMCs of the fetal ductus arteriosus have an advanced differentiation of the contractile apparatus compared with the adjacent aorta. Reexpression of fetal characteristics is seen in a number of cells in inner intima and media of the neonatal ductus arteriosus. The finding of apoptosis in these areas suggests that dedifferentiation and apoptosis are associated processes that may play a role in vascular remodeling.
Key Words: smooth muscle cells (de)differentiation ductus arteriosus aorta apoptosis
| Introduction |
|---|
|
|
|---|
-SM and
-SM actin,6 7 SM myosin 1 and 2 (SM1 and SM2),4 8 9 10 SM
-tropomyosin,11 12 h-caldesmon,8 9 13 and calponin.9 14 15 Except for
-actin, these proteins are expressed late in development and can serve as markers for the differentiated state of SMCs.8 9 10 This is also the case for an SM phosphoglucomutaserelated protein, a membrane-related protein present in adherens junctions of SMCs.16 Undifferentiated SMCs do express splice variants of fibronectin with an EDA or EDB17 and cytokeratin 8.18 19 Phenotypical undifferentiated SMCs are found not only in vessels during early development but also in the intimal layer of the vessel in the adult.8 17 18 19 20 21 22 In this study, we investigated SMC differentiation during development of intimal thickening in the human DA using several contractile proteins and EDA and EDB fibronectin as markers. The DA is a muscular artery that connects the elastic aorta and pulmonary artery during fetal life. The DA is special because it forms intimal thickening during the last two trimesters of pregnancy that can occlude the lumen when the vessel constricts postnatally. Because the contractile properties of the DA SMCs need to be optimally functional at birth, we expect that these SMCs will have an advanced development of the contractile apparatus.
Intimal thickening formation in the DA closely resembles pathological intimal thickening.23 24 It starts with lifting of the endothelial cells25 and accumulations of hyaluronan in the subendothelial region, creating an environment that is very suitable for migration of SMCs through the fragmented elastic lamina into the subendothelial region.23 This process continues until closure of the DA, providing an active center of remodeling directly beneath the endothelium.24 In this study, special attention was paid to these areas in the inner intima and to remodeling areas in the media with CN. It is suggested that CN is related to ischemia of the vessel wall and follows functional closure. CN is defined as a loss of nuclei without any accompanying cellular reaction,26 27 features that may indicate the presence of apoptosis.28 Because recent reports have shown that apoptosis contributes to arterial remodeling in the lamb during development29 and to the regulation of cellularity in experimental intimal thickening in the rat,30 we became interested in whether apoptosis of SMCs occurs in the areas of remodeling in the neonatal DA and whether this apoptosis is related to a specific SMC phenotype.
The DA thus provides a model in which SMC differentiation can be studied from the fetal to the neonatal period in both media and developing intimal thickening.
| Methods |
|---|
|
|
|---|
Two normal neonatal DAs were fixed in 4% paraformaldehyde in PBS, embedded in paraffin, sectioned at 5 µm, and used for in situ end labeling of DNA fragments. Eight other specimen were rinsed in PBS containing 6.8% sucrose, submerged in OCT compound (Tissue Tek, Miles Laboratories), quickly frozen in liquid nitrogenchilled isopentane, and stored at -20°C. Sections 5 µm thick were cut at -12°C, stored at -20°C, and used for immunohistochemistry. HE and RF staining were used to study the morphology of the vessels.
Antibodies
To study the differentiation of SMCs, we used monoclonal antibodies recognizing SM1 and SM2 (1:3),9 calponin,9 h-caldesmon,9 and SM phosphoglucomutaserelated protein.16 The intermediate filament desmin was studied with the monoclonal antibody RD 301.33
For the detection of SM
-actin, we used the monoclonal antibody 1A4 (DAKO A/S, 1:200). Splice variants of fibronectin were detected with the antibodies IST-934 35 and BC1,36 which recognize EDA and EDB sequences, respectively.
The integrity of the cytoskeleton was studied with the monoclonal antibody RV 202, which recognizes vimentin.37
Immunohistochemistry
Immunohistochemical staining was performed by a fluorescence and a peroxidase method. Just before use, the sections were fixed in cold acetone (-20°C) and were air-dried for 1 hour at room temperature. The antibody recognizing an SM phosphoglucomutaserelated protein required an additional cold methanol fixation. For the immunofluorescence method, we rehydrated the sections and rinsed them in PBS. The primary antibody was diluted, if necessary, in PBS containing 0.1% BSA and 0.05% Tween 20 and was incubated overnight at room temperature. After 3 washing steps in PBS, bound antibodies were detected with a tetramethylrhodamine isothiocyanate isomer Rconjugated rabbit anti-mouse antibody (diluted 1:50, DAKO A/S). Sections were washed again and mounted in a mixture of 80% glycerol and 20% PBS (pH 8.0) containing 1 mg/mL p-phenylenediamide, which protects the fluorescent signal from fading.
For the immunoperoxidase staining, the sections were rehydrated, rinsed in PBS, and incubated in PBS containing 0.3% H2O2 for 15 minutes to remove the endogenous peroxidase activity. Incubation of the primary antibody was identical to the fluorescence method. Bound antibodies were detected with a horseradish peroxidaseconjugated rabbit anti-mouse antibody (1:300, DAKO A/S) and exposed to 0.04% diaminobenzidine tetrahydrochloride in 0.05 mol/L Tris-maleate buffer (pH 7.6) with 0.006% H2O2 for 8 minutes. The reaction was stopped in PBS, and sections were counterstained with hematoxylin. The sections were dehydrated in graded ethanols and mounted in Entellan (Merck).
Immunohistochemical staining results were studied with a light microscope and scored by at least two independent researchers.
In Situ End Labeling of DNA Fragments
The protocol for in situ end labeling of DNA fragments was an adaptation of the technique as described by Wijsman et al.38 Briefly, paraformaldehyde-fixed tissue sections were deparaffinized, rehydrated, and incubated with citric acid (3%) for 1 hour. Tissue sections were digested with proteinase K (Boehringer Mannheim) for 30 seconds to 1 minute at room temperature to allow the enzymatic incorporation of nucleotides. Sections were rinsed in buffer A (Tris HCl, MgCl2, BSA) for 10 minutes and dried. Thereafter, sections were incubated in the same buffer containing additional nucleotides (0.01 mmol/L dATP, dCTP, dGTP [Sigma] and 0.01 mmol biotin-16-dUTP [Boehringer Mannheim]) and 20 U/mL of the Klenow fragment of DNA polymerase I (Boehringer Mannheim). Incorporated biotin-16-dUTP was demonstrated by incubation of the sections with a monoclonal antibody that recognizes biotin (DAKO A/S) at a dilution of 1:40 for 30 minutes. The antibody was visualized by a goat anti-mouse peroxidase and the chromogen 3-amino-9-ethylcarbazole (Sigma). Negative controls included omission of the Klenow fragment from the labeling mixture. Sections were counterstained with hematoxylin and mounted in resin.
Transmission Electron Microscopy
To verify the presence of apoptotic cells in the neonatal DA, a neonatal DA was fixed in half-strength Karnovsky's fixative,39 buffered in cacodylate (pH 7.2), and embedded in epoxy resin. Ultrathin sections were stained according to standard procedures with 7% uranylacetate (Merck) for 20 minutes and with lead citrate (Merck) for 10 minutes and examined in a Philips 201 electron microscope.
| Results |
|---|
|
|
|---|
|
Some of the immunohistochemical patterns of monoclonal antibodies with fetal arteries are shown in Figs 2
and 3
. Results are summarized in the Table
. Antibodies against SM
-actin (Table
) and SM-myosin (Fig 2C
and 2D
) were used to stain the complete media of the DA (Fig 2C
) and ascending (Fig 2D
) and descending aortas. Antibodies that recognized calponin (Fig 2A
and 2B
), h-caldesmon (Table
), and an SM phosphoglucomutaserelated protein (Table
) showed a stronger staining of the SMCs of the DA (Fig 2A
) compared with SMCs of the ascending (Fig 2B
) and descending aortas of the same age. Differences in staining were most significant with the antibody that recognized h-caldesmon. Expression of these proteins increased with gestational age in both the DA and the aorta. Fibronectin EDA and EDB variant expression was very low in the ascending aorta (Fig 3B
). Expression in the DA (Fig 3A
) and descending aorta was similar.
|
|
|
Neonatal Arteries
The sections stained with HE and RF showed that all neonatal DAs had well-developed intimal thickening, with an inner intima consisting of a few cell layers. Areas with CN were found in the media of 5 of 11 DAs (Figs 1
and 4
).
|
Some of the immunohistochemical staining patterns of the neonatal vessels are illustrated in Figs 5
, 6
, and 7
. The other vessels showed similar staining patterns. The results are summarized in the Table
.
|
|
|
Antibodies against SM
-actin (Table
), SM myosin (Fig 5D
), calponin (Table
), and h-caldesmon (Figs 5A
, 5B
, and 6A
) showed a strong staining of the medial SMCs and of the SMCs of the outer intima. Staining for an SM phosphoglucomutaserelated protein (Table
) was also observed, although less strong than for the other markers. Areas of CN had less staining for SM
-actin (Table
) and SM myosin (Fig 5D
), and staining was almost absent for calponin (Table
), h-caldesmon (Figs 5A
, 5B
, and 6A
), and an SM phosphoglucomutaserelated protein (Table
); this was also the case for regions in the inner intima. A contrast between these regions in the inner intima and areas of CN and the rest of the intima and normal media were most clear with h-caldesmon (Figs 5A
, 5B
, and 6A
). In the neonate, only the ascending aorta was studied. Results for the descending aorta are described by Glukhova and colleagues8 17 19 and correspond to our results for the ascending aorta. The ascending aortas studied lacked intimal thickening, as in the fetus (Fig 1
). For the late-differentiation markers, the media of the neonatal aorta (Fig 5C
and 5E
) showed staining results similar to those of the medial DA SMCs.
Fibronectin variants EDA and EDB had a high expression in the inner intima and in the areas of CN (Fig 7A
). Staining in the outer intima and media was less strong. Expression of EDA (Fig 7B
) and EDB (Table
) in the aorta was low.
Cytoskeletal integrity of the SMCs in the inner intima and areas of CN was shown by equal vimentin staining of the SMCs in these areas compared with those in the outer intima and media (Fig 6B
and 6C
). The intermediate filament desmin, a marker of muscle differentiation, did show the same staining pattern as the cytoskeletal proteins, ie, a diminished staining of areas in the inner intima and areas of cytolytic necrosis in the media compared with the outer intima and media (data not shown).
Apoptosis
With a technique of in situ end labeling of DNA fragments, a method to detect apoptosis in situ, strong staining was found in the areas of CN in the neonatal DA. Abundant apoptotic cells (>70%) were found both in areas with progressed CN and in surrounding areas with starting CN (Fig 8B
and 8C
) that had no significant loss of cell nuclei. Isolated apoptotic cells (<5%) were found in the inner intima of the DA (Fig 8A
). Outer intima, media, and adjacent aortic tissue did not show any signs of apoptosis.
|
These results were confirmed by transmission electron microscopy (Fig 8D
). With both methods, the nuclear fragmentation typical of apoptosis could be observed (Fig 8C
and 8D
).
| Discussion |
|---|
|
|
|---|
-SM actin, SM myosin, calponin, and h-caldesmon as well as the membrane-related protein SM-phosphoglucomutaserelated protein as markers. During early development, SMCs of the media of the DA are more differentiated than those of the aorta of the same age, whereas at the neonatal stage, differentiation of medial SMCs of DAs and aortas is comparable. Advanced differentiation of the DAs based on SM2 expression patterns has been described in the rabbit.40 Rabbit DAs express SM2 in a far earlier stage than the adjacent great arteries and have more intracellular myofilaments. In our human vessels, we did not find any differences for SM myosin. Probably the positivity for SM2 is overruled by the staining for SM1, because the antibody used recognizes both SM1 and SM2.9 The fetal expression pattern of EDA and EDB splice variants of fibronectin, markers of the undifferentiated phenotype,17 poses some questions. The low expression of these proteins in the ascending aorta does not correlate with the also low expression of late differentiation markers in this vessel. Maybe the expression of these markers is not fully coordinated. However, the immunostainings with cytokeratin 8, 18, and 19 antibodies (Reference 4141 and J.S., unpublished results, 1995) do fit the late differentiation markers. This special characteristic of the ascending aorta may be of interest in relation to relative specific abnormalities of the ascending aorta as observed in Marfan's syndrome.42 The descending aorta, however, does fit in the expected pattern.
In the neonatal DA, undifferentiated SMCs are found in a number of regions in the inner intima and in areas of CN in the inner media of the neonatal DA. In the areas of CN, we can be certain that these undifferentiated SMCs represent SMCs that dedifferentiate on activation, because these areas contained only differentiated SMCs before the CN started in the prenatal period. Because the same phenotypic characteristics (loss of contractile proteins, upregulation of EDA and EDB fibronectin, and apoptosis) are found in both the areas of CN and the inner intima, it is likely that a similar dedifferentiation process is happening in the inner intima as well. This inner intima is known to be a very active region in intimal cushion formation.24 Development of intimal cushions continues until closure of the DA, with subsequent lifting of endothelial cells and accumulations of hyaluronan in the subendothelial region in neonatal stages,24 most probably associated with a dedifferentiation of the migrating (or just migrated) SMCs. EDA fibronectin may play a role by accelerating the activation process, because this protein has been described to be functionally active in the repair response in wound-healing processes.43
The phenomenon of CN is not well understood. In the DA, it is thought that CN is a result of the first postnatal contractions and is characterized by a loss of nuclei, without an inflammatory response.26 27 By in situ end labeling of DNA fragments, we found extensive apoptosis in these areas of CN. Both processes, apoptosis and dedifferentiation of the SMCs, start before a loss of nuclei can be observed. For a number of regions in the inner intima, we also encountered apoptosis, although to a far lesser extent than in the CN areas. These results suggest that at least in the DA, dedifferentiation of SMCs and apoptosis are associated processes. This hypothesis is supported by data from the tobacco hawkmoth Manduca sexta.44 During apoptosis of the intersegmental muscle cells of these insects, several genes are upregulated and others are repressed. The repressed genes include actin and myosin heavy chain. Moreover, loss of
-SM actin expression and concomitant apoptosis have also been described in the intimal thickening of venous grafts.45
The observation of features similar to dedifferentiation of SMCs and the presence of apoptosis in areas of the inner intima and in areas of CN (although both areas vary in the degree of these changes) is surprising, because the situations in these regions are extremely different. As the intimal thickening develops, SMCs in the inner intima migrate, proliferate, and dedifferentiate upon stimuli present during a lasting period of fetal and neonatal stages. In contrast, in areas of CN, SMC dedifferentiation and apoptosis occur very rapidly (within a few days) and are not accompanied by proliferation and migration of SMCs.
If we would like to hypothesize on a common mechanism underlying apoptosis in the inner intima and inner media in the case of CN, this could be the process of remodeling of wall layers. The inner intima is in a constant change while actively increasing in size and restructuring into a more differentiated outer intima. The inner media remodels by CN, a process of cell loss effected primarily by apoptosis. This would fit with the role described for apoptosis in arterial remodeling in the lamb29 and in experimental intimal thickening in the rat.30
Further characterization of the intimal and medial SMC phenotypes is necessary and will be pursued by culture of the two cell types. This might give more insight into cell lineage and heterogeneity questions.
In conclusion, we found that DA SMCs have an advanced differentiation during development compared with the adjacent aorta. In neonatal stages, dedifferentiation of SMCs was observed in the inner intima and in areas of CN if present. The presence of apoptotic cells in these areas suggests that in the DA, dedifferentiation of SMCs and apoptosis are associated processes in remodeling of the vessel wall.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received July 10, 1996; accepted September 19, 1996.
| References |
|---|
|
|
|---|
-smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol. 1986;103:2787-2796.This article has been cited by other articles:
![]() |
K. N. Ivey, D. Sutcliffe, J. Richardson, R. I. Clyman, J. A. Garcia, and D. Srivastava Transcriptional Regulation During Development of the Ductus Arteriosus Circ. Res., August 15, 2008; 103(4): 388 - 395. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Agren, A. L. Cogolludo, C. G. A. Kessels, F. Perez-Vizcaino, J. G. R. De Mey, C. E. Blanco, and E. Villamor Ontogeny of chicken ductus arteriosus response to oxygen and vasoconstrictors Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2007; 292(1): R485 - R496. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Reese, J. D. Anderson, N. Brown, C. Roman, and R. I. Clyman Inhibition of cyclooxygenase isoforms in late- but not midgestation decreases contractility of the ductus arteriosus and prevents postnatal closure in mice Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2006; 291(6): R1717 - 1723. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Han, J.-K. Wen, B. Zheng, Y. Cheng, and C. Zhang Serum deprivation results in redifferentiation of human umbilical vascular smooth muscle cells Am J Physiol Cell Physiol, July 1, 2006; 291(1): C50 - C58. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Costa, S. Barogi, N. D. Socci, D. Angeloni, M. Maffei, B. Baragatti, C. Chiellini, E. Grasso, and F. Coceani Gene expression in ductus arteriosus and aorta: comparison of birth and oxygen effects Physiol Genomics, April 13, 2006; 25(2): 250 - 262. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Yokoyama, S. Minamisawa, S. Adachi-Akahane, T. Akaike, I. Naguro, K. Funakoshi, M. Iwamoto, M. Nakagome, N. Uemura, H. Hori, et al. Multiple transcripts of Ca2+ channel {alpha}1-subunits and a novel spliced variant of the {alpha}1C-subunit in rat ductus arteriosus Am J Physiol Heart Circ Physiol, April 1, 2006; 290(4): H1660 - H1670. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Nishimura, W. Li, Y. Hoshino, T. Kadohama, H. Asada, S. Ohgi, and B. E. Sumpio Role of AKT in cyclic strain-induced endothelial cell proliferation and survival Am J Physiol Cell Physiol, March 1, 2006; 290(3): C812 - C821. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Reese Death, dying, and exhaustion in the ductus arteriosus: prerequisites for permanent closure Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2006; 290(2): R357 - R358. [Full Text] [PDF] |
||||
![]() |
M. Levin, D. McCurnin, S. R. Seidner, B. Yoder, N. Waleh, S. Goldbarg, C. Roman, B. M. Liu, J. Boren, and R. I. Clyman Postnatal constriction, ATP depletion, and cell death in the mature and immature ductus arteriosus Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2006; 290(2): R359 - R364. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. K. Van Kien, F. Mathieu, L. Zhu, A. Lalande, C. Betard, M. Lathrop, F. Brunotte, J.-E. Wolf, and X. Jeunemaitre Mapping of Familial Thoracic Aortic Aneurysm/Dissection With Patent Ductus Arteriosus to 16p12.2-p13.13 Circulation, July 12, 2005; 112(2): 200 - 206. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. van der Veer, Z. Nong, C. O'Neil, B. Urquhart, D. Freeman, and J. G. Pickering Pre-B-Cell Colony-Enhancing Factor Regulates NAD+-Dependent Protein Deacetylase Activity and Promotes Vascular Smooth Muscle Cell Maturation Circ. Res., July 8, 2005; 97(1): 25 - 34. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Lambert, N. Malik, L. Shepherd, J. Gunn, S. E. Francis, A. King, D. C. Crossman, D. C. Cumberland, and C. M. Holt Localization of c-Myb and Induction of Apoptosis by Antisense Oligonucleotide c-myb After Angioplasty of Porcine Coronary Arteries Arterioscler. Thromb. Vasc. Biol., November 1, 2001; 21(11): 1727 - 1732. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Walsh, R. C. Smith, and H.-S. Kim Vascular Cell Apoptosis in Remodeling, Restenosis, and Plaque Rupture Circ. Res., August 4, 2000; 87(3): 184 - 188. [Full Text] [PDF] |
||||
![]() |
P. Neuville, M.-L. Bochaton-Piallat, and G. Gabbiani Retinoids and Arterial Smooth Muscle Cells Arterioscler. Thromb. Vasc. Biol., August 1, 2000; 20(8): 1882 - 1888. [Full Text] [PDF] |
||||
![]() |
H.-S. Kim, K.-K. Hwang, J.-W. Seo, S.-Y. Kim, B.-H. Oh, M.-M. Lee, and Y.-B. Park Apoptosis and Regulation of Bax and Bcl-X Proteins During Human Neonatal Vascular Remodeling Arterioscler. Thromb. Vasc. Biol., April 1, 2000; 20(4): 957 - 963. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. J. Hamann, J. E. Vieira, A. J. Halayko, D. Dorscheid, S. R. White, S. M. Forsythe, B. Camoretti-Mercado, K. F. Rabe, and J. Solway Fas cross-linking induces apoptosis in human airway smooth muscle cells Am J Physiol Lung Cell Mol Physiol, March 1, 2000; 278(3): L618 - L624. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. L van Heerde, S. Robert-Offerman, E. Dumont, L. Hofstra, P. A Doevendans, J. F.M Smits, M. J.A.P Daemen, and C. P.M Reutelingsperger Markers of apoptosis in cardiovascular tissues: focus on Annexin V Cardiovasc Res, February 1, 2000; 45(3): 549 - 559. |